Project Summary Allogeneic hematopoietic stem cell transplantation (HCT) is an established immune-based therapy for acute myelogenous leukemia (AML), and provides a setting for dissecting the immune basis of response and resistance to immunologic selective pressure. In particular, HCT provides an effective platform for subsequent immunomodulation to enhance graft-versus-leukemia (GvL) effects, and is thus an opportunity to develop combinatorial therapy. At DFCI, we have advanced the engineering of combinations of HCT with other immune modalities over two decades, including through phase I studies of post-HCT donor lymphocyte infusion (DLI), whole tumor cell vaccines, and checkpoint blockade antibody (CPB) therapy, by which we have evaluated the impact of the various components of these combined therapies. For example, we previously reported that patients with chronic myeloid leukemia who generated detectable marrow-infiltrating CD8+ T cells after HCT were more likely to develop durable remission to DLI, and that DLI response was associated with reversal of transcriptional signatures of T cell exhaustion, consistent with the provision of `immunologic help' (Bachireddy Blood 2014). We hypothesize that dissection of how leukemia cells and their surrounding immune cell populations co-evolve in relationship to allo-HCT course will provide essential insights for undertaking the rational design of effective combination therapy. We focus on studies of AML following HCT, as several informative clinical trials have been recently completed at DFCI (Project 1). Using modern immunogenomic tools (Core 3) and clinical factor association analysis (Core 1), we will map how leukemia and donor immune cells co-evolve following HCT in order to better strategize about the design of future studies of post-HCT immunomodulatory therapy. We will leverage our expertise in the study of clonal evolution to investigate the immunogenomic features of AML cells (i.e. neoantigen and minor histocompatibility antigen (mHAg) load, somatic mutations in antigen processing/presentation machinery) of ~200 pre-HCT leukemias (samples provided by Core 2) in relation to subsequent outcome following HCT alone, or with post-HCT vaccines, DLI or CPB (Core 1), and integrate genetic and transcriptional information from matched pre- and post-transplant relapse leukemia samples to identify the basis of immunologic escape following exposure to immune-based selective pressure (Aim 1). In parallel, we will determine the changes in the composition and functional state of marrow-infiltrating immune cells following post-HCT immunomodulation through single cell transcriptome characterization of samples collected from patients with defined response profiles (Aim 2). Finally, we will track evolving antigen-T cell interactions in association with response to post-transplant immunomodulation in which we will link the antigen specificity (i.e. predicted personal neoantigens, leukemia- associated antigens, mHAgs) to the discovered TCR sequences, and determine the cellular state of antigen- specific T cell clones over time (Aim 3).